AI in construction won’t be taking jobs, it will save lives

The robot ‘Spot’ has proven to be truly disruptive for the global construction industry, with a number of companies using it on site says Paul Wallett, Regional Director of Trimble Solutions, Middle East and India

Artificial Intelligence (AI) is today making a profound difference to our lives, even if we do not realise its overwhelming presence around us always. A lot of simple everyday actions that most of us take for granted now – from unlocking our smartphones with a single glance or touch, to summoning our favorite digital voice assistant to set a reminder, check the latest news or play songs – all have been made possible by advances in AI.

A similarly profound transformation is currently underway in several industries with a growing acceptance and use of AI-powered robots or software agents. AI was instrumental in faster development of COVID-19 vaccines – the first set of vaccines were already being tested within three months of the virus’ discovery – it would have otherwise taken several years just to get a vaccine ready. Many other industries are already heavily using AI for contact centre, factory or warehouse automation.

Even the construction industry – generally regarded as a laggard in adoption of cutting-edge technology – has surprisingly been quick to warm up to AI. The technology is particularly useful in planning and designing phases, where it allows for advanced generative design capabilities for BIM or 3D modeling. Under generative design, the computer generates several hundred-design options basis certain pre-determined goals or constraints.

The human designer is then free to choose the recommended design; or any of the others that best meet the project’s goals – and is essentially able to create the best design possible in a much shorter time. The net outcome is a higher quality design along with dramatic savings in designer’s time – which is then available for use in other projects.

When it comes to onsite construction, modern safety, monitoring, and maintenance systems are today using advanced AI capabilities to automatically predict and warn supervisors about adverse safety situations. In doing so, these systems minimise the role of human error in any adverse incident.

Suffolk, a major US-based construction company, has been developing a system which could potentially predict and therefore prevent accidents before they happen. The system incorporates a form of AI called deep learning and image-recognition software, along with cameras that routinely take shots of ongoing work throughout an active construction site. The AI-based system then analyses these images and compares what it ‘sees’ to how things ‘should or should not be’, using historic data collected over a decade. When put to work for monitoring a new construction site, it can quickly flag situations that seem likely to lead to an accident, such as workers not wearing safety equipment or working too close to a dangerous part of a machinery.

Similar tools incorporating AI are also able to track the real-time interactions of workers, machinery and other objects on the site, and can then alert supervisors or offsite managers of construction errors, productivity issues or willful ignorance of safety protocols. A couple of years back, Trimble announced a partnership with Hilti and Boston Dynamics to utilise robots for routine tasks in hazardous environments to improve safety, efficiency, and data capture consistency. This robot ‘Spot’ has proven to be truly disruptive for the global construction industry, with a number of companies using it on site.

With such a wide array of applications and use cases, AI is set to dramatically transform the engineering and construction industries in the months and years to come. A natural corollary of this transformation is the fear and concern over job losses – after all, the construction sector is the second largest employer in India after agriculture – and provides direct or indirect livelihood to nearly 100mn people.

To be fair, technology-driven societal changes like what we’re experiencing with AI and automation always engender concern and fear—and for good reason. Over 200 years ago, a group of textile workers started a movement to oppose any form of machinery or automation, based on their perceived fear of job loss. They came to be known as Luddites, and even today, any person who is afraid of or unwilling to use new technology is often called a luddite.

It is important here to point out that the fears of modern-day Luddites are not completely unfounded. A McKinsey report suggested that by 2030, intelligent agents and robots could replace as much as 30% of the world’s current human labor, and ‘automation will displace between 400 and 800mn jobs by 2030’. However, it is also an undeniable truth that every major technological shift has ended up creating more jobs than what were destroyed. This is because jobs or livelihoods never get obsolete; only specific skills get obsolete and are no longer sufficient to earn a livelihood. In the 1800s, the handloom weaver was no longer needed in the modern textile factory but he could get trained on a power loom and still earn his livelihood.

Particularly in the context of construction, AI’s impact on improving workplace safety and productivity is likely to be several orders of magnitude larger than its potential impact on job losses. For perspective, let’s consider that 38 construction site deaths are reported every day in India; which is 20 times higher than in Britain. India also has the world’s highest accident rate among construction workers, according to a recent study by the International Labour Organization (ILO). Even in a developed country like the United States, the construction industry regularly records five times as many fatal accidents than any other industry, despite safety protocols and mandatory usage of protective equipment.

Construction remains one of the most dangerous jobs out there – and majority of incidents have human error to blame. Tools such as Trimble CrewSight can ensure that only those workers who are trained, qualified, and properly equipped for a given task are allowed to enter a jobsite. Embedded with AI and video-based monitoring, such tools can cut out the role of human error altogether in safety incidents, creating a win-win for all stakeholders.

Ultimately, despite widespread concerns, AI or robots are unlikely to replace humans altogether on construction sites in the near future. On the other hand, it will definitely help reduce worksite injuries and accidents, or expensive errors, and also make operations far more efficient. Leaders at construction companies should prioritise investment based on areas where AI can have the maximum impact on their company’s operations. We believe that early movers will not only generate tremendous business value for themselves both in short and long term but will also set the direction of the industry for a complete AI-led transformation.